1 code implementation • 21 Apr 2024 • Feiqi Cao, Caren Han, Hyunsuk Chung
In this work, we propose a novel tree-based explanation technique, PEACH (Pretrained-embedding Explanation Across Contextual and Hierarchical Structure), that can explain how text-based documents are classified by using any pretrained contextual embeddings in a tree-based human-interpretable manner.
no code implementations • 13 Apr 2023 • Yihao Ding, Siwen Luo, Hyunsuk Chung, Soyeon Caren Han
Document-based Visual Question Answering examines the document understanding of document images in conditions of natural language questions.
1 code implementation • 4 Apr 2023 • Yihao Ding, Siqu Long, Jiabin Huang, Kaixuan Ren, Xingxiang Luo, Hyunsuk Chung, Soyeon Caren Han
Compared to general document analysis tasks, form document structure understanding and retrieval are challenging.
no code implementations • 29 Nov 2022 • Zhihao Zhang, Siwen Luo, Junyi Chen, Sijia Lai, Siqu Long, Hyunsuk Chung, Soyeon Caren Han
We propose a PiggyBack, a Visual Question Answering platform that allows users to apply the state-of-the-art visual-language pretrained models easily.
no code implementations • 27 May 2022 • Yihao Ding, Zhe Huang, Runlin Wang, Yanhang Zhang, Xianru Chen, Yuzhong Ma, Hyunsuk Chung, Soyeon Caren Han
We propose V-Doc, a question-answering tool using document images and PDF, mainly for researchers and general non-deep learning experts looking to generate, process, and understand the document visual question answering tasks.
no code implementations • CVPR 2022 • Yihao Ding, Zhe Huang, Runlin Wang, Yanhang Zhang, Xianru Chen, Yuzhong Ma, Hyunsuk Chung, Soyeon Caren Han
We propose V-Doc, a question-answering tool using document images and PDF, mainly for researchers and general non-deep learning experts looking to generate, process, and understand the document visual question answering tasks.